JOURNAL OF COMPUTATIONAL METHODS IN ENGINEERING (ESTEGHLAL)
Year:2011 | Volume:29 | Issue:2|Papers Count:8

In this paper, the short channel effects of different gate configurations and geometry parameters of carbon nanotube (CNT) field-effect transistors with doped source and drain extensions are investigated. The simulation is based on the self-consistent solution of the three-dimensional Poisson equation and Schrödinger equation with open boundary conditions, within the nonequilibrium Green’s function formalism. Simulation results show that double gate structure offers quasi-ideal subthreshold slope and drain induced barrier lowering even for the rather thick oxide (5nm). Then, the investigation of electrical characteristics of double gate carbon nanotube field-effect transistor (DG-CNTFET) shows that as the CNT normalized density or CNT diameter increases, the current in the on-state increases as well. Also, the off-state current in DG-CNTFET decreases with increasing drain voltage. Furthermore, in the negative gate voltages, for a large drain voltage, increasing in drain current due to band to band tunneling requires a larger negative gate voltage, and for low drain voltage, resonant states appear.

The term "large panels" refers to structures that use large precast panels as their main horizontal load resistance system. Precast concrete large panels are most economical to use because of their higher quality and the reduced construction time associated with them. The most important requirement for their construction is the structural analysis of their true behavior in the structure. Nowadays, the most important part of seismic design of structures is carried out based on equivalent static method and calculation of seismic forces based on the linear spectrum of an earthquake using a structural reduction factor. The determination of the reduction factor seems necessary in structural seismic design given its utmost significance. In this research, many kinds of precast structures with different stories and bays are considered. The seismic behaviour of such structures is strongly dependent upon the characteristics of both horizontal and vertical connections between panels. A variety of assumptions have been made in this study drawing upon experimental research in which the static pushover analysis has been used. The results show that the reduction factors based on existing codes are conservative.

Due to the need for greater depth of investigation in petroleum and geothermal exploration and the complexity of the associated geological models often involved in any interpretation of magneto telluric data, it would be inevitable to employ a more robust and effective 3-D forward modeling engine capable of incorporating complex geological models. The newly developed algorithm presented in this paper uses the conventional finite difference technique of numerical methods but in a nonstaggered scheme which enables solving the governing electromagnetic fields with the imposed boundary conditions simultaneously. Through dividing the air and earth half space into discrete blocky model the governing magneto telluric equations are discretized to form a linear matrix equation which in turn is solved for the unknown magnetic and electric fields iteratively. Because of the frequency dependency of governing magnetic and electric fields and the finite size of the discretized model a number of problems such as oscillating phenomena in the final forward response in the course of solving such boundary value problems are observed. Also it happens that due to the conventional staggered scheme of forming discretized linear equations, a pseudo singularity appears which makes the iterative solvers to diverge. In order to address and remedy for these problems we have introduced an optimum mesh refining scheme to resolve the response oscillation problem. Also through using non staggered formalism which changes the shape of the final discretized matrix equation, it is shown that the pseudo singularity effect could be removed effectively.

The present paper is to examine European rollover standard (ECE66) on O457 bus, using computer simulation method. Due to the huge size of the analyzed structure and unavailability of supercomputers and also according to the rollover regulation, the equivalent method, i.e., body section rollover test is selected. Also, because of impossibility of real test, the simulation results have been qualitatively compared with the existing experimental results. The extracted results show that deformation occurs mainly in the pillar-to-floor and pillar-to-roof joints, while the roof and chassis have small deformations. As the deformations in the body were greater than the standard level and the bus failed the test, suggestions were made to increase the body strength. The suggestions are based on the available capabilities and limitations of the production process.

In this paper, the effect of rotor cross-section geometry in viscous micro pumps is studied for optimizing the flow and maximizing the exit flow rate using the Overset Grid method. The main construction of viscous micro pumps consists of a rectangular channel and a rotor with circular cross-section which rotates eccentrically and perpendicular to the channel axis. In this paper, the effects of channel height, rotor eccentricity and channel pressure difference on the average velocity at the outlet of micro channel in the presence of circular, square, rectangular and elliptic rotor cross-sections are investigated. The flow equations are solved two-dimensionally, using incompressible finite volume method. The obtained results show that the maximum value of the exit average velocity is obtained at a micro channel of 1.5 times of the rotor diameter. However, increasing the rotor eccentricity and the angular velocity of the rotor causes the exit velocity magnitude in the micro pump to increase. The obtained results show very good agreement with existing experimental and numerical results.

In this paper, fully developed flow and heat transfer of viscoelastic fluid in a curved pipe with square cross section is investigated. To that end, Criminale-Eriksen-Filbey (CEF) constitutive equation is used as viscoelastic model and heat transfer is studied at constant heat flux and in constant temperature conditions. Here, the governing equations are discrete using finite difference method. Also, staggered grids are used as mesh, and Marker and Cell method is applied for allocating the flow parameters on staggered grids. Here, the effect of the stress fields’ work of viscoelastic fluid and Brinkman number on convective heat transfer inside a curved duct is studied for the first time. Overall, the inverse effect of the first and second normal stress differences on secondary flows intensity and heat transfer is the main result of the current research.

This paper addresses modeling and optimization of loading path in T-shape hydroforming of tubes. A set of experimental data designed by DOE method is used to assess the influence of loading process parameters in hydro formed geometry. The minimum thickness and maximum height of protrusion are used as formability criteria and the regression modeling is used in order to establish the relationship between the input and output parameters. The adequacy of the model is evaluated using analysis of variance technique. The proposed model is embedded into a Simulated Annealing algorithm to optimize the loading process parameters and to find the best input variables to produce T shaped tubes. The comparison between the current results and the experimental data show the capability of the method.